Upload
amado-valero
View
20
Download
0
Embed Size (px)
DESCRIPTION
Aplicaciones de instrumentos de radar en la medición de variables del estado del mar.
Citation preview
REMOTE SENSING OF SEA STATE BY RADAR
Donald E. Barrickx
Electromagnetics Division B a t t e l l e , Columbus Laboratories
Columbus, Ohio 43201
Abstract
In recent years several radar techniques have evolved which allow the remote measurement of certain parameters important in the descrip- t i o n of s e a s t a t e . A t MF and HF, monostatic and b is ta t ic conf igura t ions employing s a t e l - l i tes , ships , i s lands, and/or land based s t a t i o n s can measure the ocean waveheight spectrum with several f requencies via f i rs t - order Bragg s c a t t e r . A t high HF and VHF, the ocean waveheight spectrum can b e estimated a t a single carrier frequency via secord-order mechanisms; this technique i s espec ia l ly s u i t e d t o remote sensing via long distance ionospheric propagation. A t UHF, i t i s pos- s i b l e t o measure the slope spectrum of the longer ocean waves via cross-correlat ion of simultaneous Bragg-effect returns a t two frequencies. The short-pulse microwave sa t e l l i t e a l t ime te r pe rmi t s a d i r e c t measure- ment of the significant waveheight of the sea a t t he suborb i t a l po in t v i a t he specu la r point mechanism. Such techniques w i l l be i m - portant both for detailed oceanographic study of ocean wave c h a r a c t e r i s t i c s and for rou t ine monitoring of sea s ta te for mari t ime/meteoro- logical purposes.
Introduct ion
This paper examines several radar concepts for remotely sensing sea s ta te . "Remote" as appl ied to the concepts can mean a s c lose a s 10-20 nmi; other concepts--if implemented--could measure s e a s t a t e from a land-based s i t e a s f a r away as 2000 nmi. S a t e l l i t e and airborne sensor techniques are a lso included. Rather than present ing de ta i l ed t heo re t i ca l de r iva t ions of e lec t ro- magnetic wave s c a t t e r from the rough sea, an attempt i s made here to discuss only those remote sensing concepts for which a clear physical understanding is ava i lab le . Only when the i n t e rac t ion mechanism is understood to the point that an important c h a r a c t e r i s t i c of s e a s t a t e r e l a t e s c l e a r l y and d i r e c t l y t o a simple radar observable
5~ After August, the author w i l l b e a f f i l i a t e d wi th t he Wave Propagation Laboratory, Environ- ment Research Laboratories of the National Oceanic and Atmospheric Administration, Boulder, Colorado 80302.
can one reasonably expect success as a r e - mote sensing concept. Rarely do mathematical manipulations lead to any significant remote sensing discovery. After a con- cept is uncovered and the expected physical mechanism ident i f ied, mathematical analysis can provide a va luable quant i ta t ive bas i s f o r f u r t h e r development and design of the technique. Experimental verification i s of course ultimately necessary to demonstrate f e a s i b i l i t y and accuracy.
Four interaction concepts for remotely sensing sea s ta te are suggested here: (1) Firs t -order Bragg s c a t t e r a t MF/HP; (2) Second-order Bragg s c a t t e r a t HF/VHF; (3) Two-frequency c o r r e l a t i o n a t UHF; ( 4 ) Short-pulse alt imetry a t X-band. Of the four concepts, (1) and ( 4 ) -have been ana lyzed i n de t a i l t heo re t i ca l ly and v e r i - f ied in severa l conf igura t ions exper imenta l ly . Concept ( 2 ) has been partially analyzed theore t ica l ly , and t e s t s a r e j u s t b e i n g in i t ia ted to p rovide exper imenta l ver i f ica t ion . Only p a r t i a l t h e o r e t i c a l a n a l y s i s i s present ly available to support concept (3). A l l four w i l l be b r ie f ly d i scussed in the fo l lowing sections. Rather than detailed mathematical der iva t ions , the p r inc ipa l equa t ion which quant i ta t ively descr ibes the scat ter concept w i l l be given and explained. When experi- mental ver i f icat ion is ava i lab le , i t w i l l be presented.
The "radar range equation" describes the power received in terms of t ransmit ted power, PT, wavelength, h , ranges from t ransmi t te r to t a rge t (RT) and t a r g e t t o receiver (RR), and antenna gains. When the t a rge t is a patch of sea of a rea d A , the average radar cross section describing s c a t t e r is o = u dA, where 0 0 is the average radar cross sect ion per uni t area (dimensionless) for the sea. This equation then becomes
where FZ represents all losses g rea te r than the free-space spreading loss (e .g . , surface- wave losses, ionospheric absorption, attenua- t ion through ra in , e tc . ) . For lossless transmission, F C -f 1. For surface-wave o r l ine-of-s ight radars , GT and GR are def ined
186-OCEAN '72
h e r e t d be the equivalent t reesspace gains . of the antennas in t he d i r ec t ion of the scat ter ing patcn. For ionospheric over-the- horizon propagation, these gains are as measured in the presence of the ground ( i .e . , about 6 dB higher than free-space gains.)
It is essent ia l tha t the t e rm "sea s ta te" be de f ined a t t h i s po in t and the s ignif icant observable parameters describing it be dis- cussed. Sea s t a t e i n a p rac t i ca l s ense r e fe r s t o t he he igh t of the waves o r roughness present on the surface of the ocean. Significant Wave- height (Hv~) is the maritime descr iptor giving the height (peak-to-trough) of the highest 113 of the waves; it is roughly re la ted to the rms waveheight, h, by H1/3 - 2.83 h. A s winds d r ive the seas h'igher, they i n essence increase the heights of longer , fas te r mbving waves; t he sho r t e r waves a r e f u l l y developed t o t h e i r maximum heights . A deep-water wave of length L t r ave l s a t ve loc i ty v = s, where g = 9.81 ms-2 is the accelerat ion of gravi ty . These wind-driven waves w i l l move predominantly with the winds. The length of the longest ocean wave which the wind can exci te is one whose phase v e l o c i t y , v , j u s t matches the wind speed, u. Thus the crudest--but perhaps most important--descriptor of sea state is wave- height (HI, o r h ) . When the sea is f u l l y developed 6 the winds, a rough estimate of h i n terms of wind speed, u, is h = .016ua m, where u is i n m / s . A more quan t i t a t ive measure is the ocean waveheight spectrum, S(X), where n = 2n/L is t h e s p a t i a l wave- number. The most detailed function des- cribing the strengths of ocean waves moving i n any direct ion, 8, i s t h e d i r e c t i o n a l spectrum S(+, 9) = S ( X cos 9, u s i n e). See Kinsman [ 11 or Barr ick [ 23 f o r a d i s - cussion of ocean wave c h a r a c t e r i s t i c s .
MF/HF Bragg Sca t t e r
Nearly two decades ago, Crombie [3] experi- mentally discovered the mechanism giving rise to radar sea sca t te r a t HF by observing the Doppler spectrum of his received signal. It consisted almost entirely of two d i s c r e t e l i n e s s h i f t e d above and below the ca r r i e r ( f o = c/h) by an amount d'm Hz. For h i s ground-wave backscatter configuration, these Doppler s h i f t s were seen to be produced by ocean waves whose lengths were one-half the .'adio wavelength (L = h/2) moving toward and
away from the radar . Thus the mechanism he deduced is Bragg s c a t t e r . Of a l l t h e ocean waves present, the only ones seen by the radar a re t hose forming a d i f f r ac t ion g ra t ing w i th half-wavelength spacing, because in this case the r e tu rns from each wave c r e s t w i l l r e in fo rce coherent ly in the backsca t te r d i rec t ion . La ter theore t ica l ana lyses [ 4,5] confirmed the correctness of Crombie's deductions; Barrick [5 ] showed tha t t he magnitude of the average scattered signal spectrum, a(w), and normalized c ross s ec t ion u o (ao = [Eo (w)dw) f o r ground-wave . .
backsca t te r wi th ver t ica l po lar iza t ion is*
where = 2n/X and S ( s , 3, w)/S(%, 5) are the d i rec t iona l spatial-temporal/spatial spectra of the ocean waveheight, normalized such that,
The above equation shows Bragg s c a t t e r t o expla in the in te rac t ion mechanism, s ince the ocean wavenumber, It, which is being observed i s 2% (i .e. , L = h / 2 ) . The s t r eng th of t he s igna l depends upon the height of the waves in the spec t rum a t t h i s wavenumber. For a Phillips spectrum model ( i . e . , S ( u ) = .005/ ( 2 n 1 t ~ ) f o r f u l l y developed waves), this gives rise t o u o = 0.02 and u(w) = 0.04 $(w - w0 i 2 n m ) fo r f i r s t -o rde r waves. The impulse functions equally spaced about the carrier explain the discrete Doppler shifts observed experimentally. The value u o = .02 = -17 dB i s in quant i ta t ive agreement with experimental observat ions a lso [ 21.
Ongoing experimental confirmation and develop- ment of sensing techniques based upon t h i s concept have been conducted by Crombie and are discussed in his paper a t this meet ing [ 61. Rather than qepeating his experimental observa- t ions here , the reader is re fer red to h i s paper . Instead, w e w i l l b r ie f ly descr ibe severa l re- mote sensing configurations here which are based upon t h i s p r i n c i p l e . More .detai led analyses of these are available elsewhere [ 2 , 6 ] .
A s t a t iona ry ground-based backscatter radar which can employ seve ra l carrier frequencies i n t h e lower HF and upper MF region can provide estimates of the nondirectional wave- height spectrum, as shown i n Crombie's Fig. 2 [ 61. By loca t ing t h i s backsca t t e r r ada r on a moving sh ip , t he sh ip ve loc i ty imparts a Doppler b i a s to the signal spectrum which can permit measure- ment of the directional waveheight spectrum [ 21 . By placing a t r ansmi t t e r on a buoy ( o r ship) and a r e c e i v e r b i s t a t i c a l l y on a non-synchronons satel l i te , the directional waveheight spectrum near the buoy can be obtained [ 21 . The d i r e c t i o n a l waveheight spectrum ca: Is0 be obtained from bi- s t a t i c a l l y s e p a r a t e d ground-based t r ansmi t t e r and receiver configurations [ 2 ] ; this has been veri- f ied experimentally and reported in Peterson e t . a l .
. [71
HF/VHF Second-Order Bragg S c a t t e r
The first-order theory discussed above shows that the received radar Doppler spectrum from t h e sea should consist of two narrow spikes (impulse functions), and nothing else. The dominance of t he two spikes observed in experi-
* For o the r po la r i za t ions and b i s t a t i c direct ions, see Barr ick 121.
OCEAN '72-187
mental records below about 5 MHz i s s t r i k i n g ; note Fig. I of Crombie [61, made a t 2.9 MHz. In the higher HF and VHF region, however, two e f f e c t s become evident upon examination of measured Doppler spec t r a : (1 ) t he f i r s t -o rde r "spikes" becomes broader, and (2) a l a r g e r , continuous "floor" between and near the spikes i s seen t o exist, of ten cons is t ing of peaks near the f i rs t -order spikes . It was suggested by Hasselmann [ 81 and Barrick [ 9,2] that these second-order peaks near the first-order spikes should in themselves be measures of the ocean waveheight spectrum.
A recent example of such a signal spectrum is shown in F ig . 1; t h i s was measured a t 30 MHz with a ground-wave backsca t te r radar (ver t ica l ly polar ized) by Tyler et . a l . [ l o ] . S p e c t r a l
I I I I I I I -1.5 - I O -0.5 0 0 5 1.0 15 2 0 2 5
,Normollzed Doppler Shtfl g/r x - f-fo
Fig. 1. Sea echo Doppler spectrum a t 30 MHz f o r near-grazing backscat ter , ver t ical polar izat ion: January 25, 1972 [Af te r Tyler , e t . a l . , 101.
process ing be t te r than 0,001 Hz was obtained, and some spa t ia l / spec t ra l averaging was done t o smooth the record . Whi le quant i ta t ive de ta i l s of the actual ocean waveheights in the measure- ment area were not available, the winds were observed to be fa i r ly high and toward the radar , r e s u l t i n g i n f a i r l y rough seas. The f i r s t - o r d e r Bragg l i n e s are evident--very near their pre- dicted posi t ions of 0.56 Hz; land echo appears as a narrow sp ike a t zero Doppler. The presense of two peaks near the f irst-order echoes is qui te ev ident ; these peaks merge i n t o a continuum, which ro l l s o f f to the sys tem noise l eve l above and below t h e Bragg spikes. Tyler, et . a l , observe that for this record, about one-half the to ta l sea echo power appears in these second- order sidebands.
A f u r t h e r example of such a sea-scatter spectrum made via ionospheric propagation is shown i n Fig. 2. This was measured by Barnum (see B a r r i c k [ 2 ] ) a t 25.75 MHz a t a distance of 2700 km from the radar in southern Cal i forn ia . Barnum's equipment permitted a spec t r a l r e so lu t ion of 0.04 Hz, and some spa t i a l ave rag ing was done t o smooth the record. In the i l luminated area of the Pacific, winds (and waves) were predominantly toward t h e west (away from the radar), producing higher negative Dopplers. The broadened f i r s t - order Doppler sp ikes , p red ic t ed t o be 3=0.518 Hz from the re turn ing carrier, are c lear ly ev ident . Second-order spikes are again present . Some spectral smearing due to ionospheric motions is
inevi table with such sky-wave sea-scat ter records; there is a l so a small percentage of the time when disturbed ionospheric conditions preclude such measurements e n t i r e l y .
0
-10
< -20 5 L. a
a v) '0 0
01 u Q,
-30 .-
a .- + 0 -40 Q, a
-50
-60
I
I I 1 I '1 I I '1
rlormalized values
I 1 I I si I I II
-1.0 - 0.5
I shift = + 0.07 Hz
i I J A
0.5 I .o Doppler Frequency, Hz
Fig. 2. Sea echo Doppler spectrum a t 25.75 MHz f o r ionospherically propagated backscatter: March 30, 1971 [After Barnum, see [2] 1 .
Barrick [ 21 has shown theo re t i ca l ly t ha t bo th hydrodynamic and electromagnetic second-order effects produce the peaks and continuum near t he f i r s t -o rde r spec t r a l sp ikes . An i n t e g r a l representat ion for the average backscat tered spectrum, o(v), ( 7 = w - tuo) can be wri t ten in terms of the first-order directional waveheight spectrum S ( K ) = S ( 5 , M. ) as follows: Y
A ground-wave geometry i s assumed, with propagation along the x-axis. The kernal of t he i n t eg ra l , r(Zl, K ~ ) , accounts for both electromagnetic and hydrodynamic e f f e c t s . The former are obtained from the second-order terms of the Rice boundary pertur- bat ion theory [2] , and indicate a double-bounce Bragg-reflection mechanism. The hydrodynamic con-
-
188-OCEAN '72
varied with the s lope of the longer gravi ty waves passing through the range gate. Spectrum analysis of this ampli tude w i l l indirect ly yield the s lope spectrum of the longer "sea state" waves. This slope spectrum is of course direct ly re la ted to the waveheight spectrum by a factor consisting of the square of t he wavenumber. The short-pulse technique mentioned here was examined experimentally and theoretically by Sovie t inves t iga tors [ 111.
An al ternat ive to the short-pulse experiment discussed above can provide the same information--based upon the same mechanism--but eliminates the need f o r a Fourier transform of the received signal envelope. In this experiment , two frequencies are simultaneously transmitted. This can be accomplished by a balanced modulator a t t h e t r a n s m i t t e r o u t p u t . The frequency separat ion, Af = f, - f l , r e s u l t s i n a "Separation wavenumber, It A&. = 2nllfIc. While scatter a t t h e two UHF frequencies, f , and f,, takes place via the f ami l i a r Bragg mechanism, the slopes of the longer, underlying gravity waves cause the scat tered power a t the two frequencies to become less c o r r e l a t e d a s the f requencies are more widely separated. Results of an ana lys i s of th i s t echnique [2 , 121 show t h a t the covariance of the received power a t t h e two frequencies for a one-dimensionally rough surface model a t a given backscatter angle, 8, from the v e r t i c a l is expressible as
Var [P(Ak)] = K1 + K, WsL (2Ah s i n 8) , ( 4 ) where K1 and K, are constants (functions of the geome- t r y and the pu lse i l lumina t ion pa t te rn) . WSL(U) is the one-dimensional slope spectrum of the longer g rav i ty waves. Equation ( 4 ) shows a s u r p r i s i n g r e s u l t ; the slopes of the longer gravity waves appear to be measured by a Bragg-scatter process occurring a t carrier frequency Af, producing a sampling s p a t i a l wavenumber 2Ah s i n 8. Thus the frequency separation, Af, should be varied between about 1.5 and 15 MHz t o measure t h e wave spectrum of the longer "sea s t a t e " waves.
This technique, while as yet untested experimentally, has many potent ia l advantages, pr imari ly in equip- ment s i m p l i c i t i e s . Two frequencies with less than 2 percent maximum sepa ra t ion i n terms of the car r ie r are e a s i e r t o g e n e r a t e i n a quasi-CW manner than implementation of a short-pulse system with a spectrum analyzer a t the rece iver ou tput . This system could ul t imately f ind appl icat ion in a i rborne o r sa te l l i t e p la t forms , e i ther separa te ly o r in b i - s ta t ic conjunction with shipboard receivers.
Short-Pulse Radar Altimetry
Cons iderable in te res t in the pas t two years has turned to another remote-sensing technique for ocean prof i le observat ions: the short-pulse sa t e l l i t e -bo rne microwave radar altimeter. I n i t i a l l y such experiments were conceived for geodetic purposes, which could measure the in- stantaneous mean sea level t o a precision of about 10 cm, thus obtaining maps of the geoid and slopes of gross oceanic trenches. It was recognized, how- ever, that waveheight of the order of twenty f e e t o r more would s t r e t c h a short pulse (e.g., one
whose e f fec t ive l ength is of the order of a f o o t ) , necess i t a t ing ca re fu l ana lys i s and in t e rp re t a t ion t o e x t r a c t t h e mean surface posi t ion a t the suborbi- t a l point from the sea-dis tor ted s ignal . A byproduct of t h i s processing--which t o some presently appears t o be a more s ign i f i can t u se of the a l t imeter than maps of the mean sur face p rof i le - - i s the s ign i f icant waveheight of the ocean waves below t h e s a t e l l i t e .
The in t e rac t ion mechanism between the pulse of microwave energy and the rough sea i n t h i s c a s e is Bragg scatter. A radar s igna l whose wavelength i s short compared t o t h e l a r g e r dimen- sions of the ocean waves is backscattered from the reg ion near the ver t ica l (i.e., the suborb i ta l region) via the specular point mechanism. This means that only those wave f ace t s whose normals .. point toward the radar produce scatter. This i s the same mechanism as the danc ing g l i t t e r po in ts one observes visual ly due t o t h e b i s t a t i c s c a t t e r of the sun or moon o f f a rough lake in the reg ion near the specular direct ion. The r e f l e c t i o n from each specular (or g l i t t e r ) po in t is proportional i n s t r e n g t h t o t h e s u r f a c e r a d i i of c u r v a t u r e a t t ha t po in t , and the s igna l s from various points combine incoherently due t o random phase differences between them.
The analysis of the sea-scattered signal observed by a short-pulse a l t imeter via the specular point theory (i.e., e i t h e r a geometrical o r physical opt ics formulation) has been performed elsewhere. . Miller and Hayne[13]used a simple model for specular face t s ca t t e r ; t he r eade r is referred to Barr ick [2] for a s t ra ight-forward der ivat ion in terms of the complete specular point theory and simple Gaussian models for the radar pat terns . This leads to the fol lowing closed-form solution for the product of the radar c ross sec t ion times the antenna gains--as a function of time--which appears i n Equation (1) f o r t h e re- ceived altimeter power:
where H = a l t imeter he ight ; s = rms surface s lope ; 7 = half-power pulse width ; s NN . 0 7 4 G i n
terms of wind speed, m / s
~i~ = one-way half-power antenna beamwidth ; x w = C T / ( 4 m ) ; t p = 2 J g + 2h21c ;
and h is t h e rms waveheight. The funct ion 9 is the e r ror func t ion . I t is assumed here that t h e altimeter beam i s poin ted d i rec t ly toward the v e r t i c a l and t h a t a n unskewed Gaussian function represents the he ight d i s t r ibu t ion of t he ocean waves.
The constants tp and ts have interest ing physical i n t e rp re t a t ions : tP is the effect ive length of the s t re tched pulse after s c a t t e r from the waves d is t r ibu ted over a plane surface, while ts i s the ( temporal) depth of the effect ive scat ter ing region on the spher ica l ear th sur face . I f ts >> t,, the
190-OCEAN '72
altimeter is sa id to be pu lse- l imi ted in iti opera- t ion; the opposi te extreme is c a l l e d beam-limited operation (see Barrick [2] for more de t a i l s ) . Sa t e l - l i t e a l t ime te r s a r e nea r ly always pulse-limited be- cause o f the i r a l t i tude and antenna sizes. In such operat ion, the s lope of the leading edge of t he re- tu rn is influenced by sea state . This is i l l u s t r a t e d by predict ions from the model in F ig . 5 fo r a s a t e l - a t 435 km, wi th a beanwidth of 3', and whose e f f ec t ive pulse width i s less than 15 ns.
.0.4
. 0.3
. 0.2
. 0.1
I I I I I I I I I -90-80-70-60-50-40-30-20 -10 0 IO 20 30 40 x) 60 70 BO 90
Time, nanoseconds
Fig. 5. Leading edge of averaged altimeter output versus t ime predicted for pulse-limited operation.
Limited experimental confirmation is ava i l ab le from aircraf t quasi-pulse l imited a l t imeter operat ion. Two such records made by Raytheon a re shown h e r e i n Fig. 6 (see Barr ick [2] for detai ls) . Surface winds i n each case were measured a t -12 and 22 knots. From the rise times of the s ignals , the wind speeds infer red from Equation (5), along with the dependence of waveheight upon wind speed given in the In t roduct ion , are 14.1 and 21.2 knots. Thus prel iminary ver i f ica- t i o n of t he model is encouraging.
-I-
-I F tr=21nsec
I Flight *I4 Run *I2 H = IO kft r=20!onSec t, = 21 nSeC
Calculated wind = 1 4 . 1 knots Measured wind = E knofs
flight *I6 Run *9 H- IO kft
t,=mnSec ~=20nsec
Calculated wind =21.2 knots Measured wind = 22 knots
c I Fig. 6. Measured a i r c ra f t a l t ime te r r e sponses . Wind
speeds inferred from rise times a r e compared t o observed wind speeds.
Sumnary
Of the concepts presented here, the MF/HF f i r s t - order Bragg sca t te r t echniques a re the most advanced, bo th ana ly t ica l ly and experimentally; they are per- haps t h e most l i m i t e d i n long-term potent ia l because of the required wide frequency band of operat ion and l imited remote-sensing coverage area per station, The second-order HF/VHF concept shows g rea t po ten t i a l f o r land-based distant sensing of large ocean areas via ionospheric propagation. The UHF two-frequency technique must be developed fur ther to def ini te ly e s t ab l i sh i ts f e a s i b i l i t y ; i n a i r b o r n e o r s a t e l l i t e - sh ip appl ica t ions , it could provide several important pieces of data about sea s ta te . Short-pulse satel- l i t e a l t ime t ry i s proven experimentally; while pro- viding only the significant waveheight along the orb i ta l p lane , it could perhaps be the f i r s t t ech- nique to reach the operational stage, providing needed sea-state information over large ocean areas on a da i ly bas i s .
References
B. Kinsman, Wind Waves, Englewood C l i f f s , New Jersey: Prent ice-Hal l , Inc. , 1965.
D. E . Barr ick, "Remote sensing of sea state by radar" , in Remote Sensing of the Troposphere, V. E . Derr, Ed. Washington, D.C.: U . S . Government Pr in t ing Off ice , 1972, Chapter 12.
D. D. Crombie, "Doppler spectrum of sea echo a t 13.56 Mc/s", Nature, vol. 175, pp. 681-682, 1955.
J. R. Wait, "Theory of HF ground wave backscat ter from sea waves", J. Geophys. Res., V O ~ . 71 , pp. 4839-4842, 1966. D. E. Barrick, "First-order theory and ana lys i s of MF/H!?/VHF s c a t t e r from the sea" , IEEE Trans. Antennas Propagat. , V O ~ . AP-20, pp. 2-10, 1972.
D . D . Crombie , "Resonant backscat ter from t h e sea and i ts app l i ca t ion t o physical oceanography", Conference Record, 1972 IEEE International Conference on Engineering in the Ocean Environment, (Newport, Rhode Is land, Sept . 13-15, 1972).
A . M. Peterson, C . C. Teague, and G . L. Tyler, "Bistatic radar observation of l onppe r iod d i r ec t iona l ocean-wave sDectra w i t h LORAN A", Science, vol. 170, pp. 158-162, 1970.
K. Hasselmann, "Determination of ocean wave spec t ra from Doppler r e tu rn from the sea surfacell, Nature- Physical Science , V O ~ . 229, pp. 16-17, 1971.
OCEAN '72-191
[9] D. E. Barrick, "Dependence of second- order Doppler sidebands in HF sea echo upon sea state", in 1971 G-AP Inter- national Symposium Digest, (Los Angeles, Calif., Sept. 21-24, 1971), pp. 194-197, 1971.
[lo] G. L. Tyler, W. E. Faulkerson, A. M. Peterson, and C. C. Teague, "Second-order scattering from the sea: Ten meter wave- length observations of Doppler continuum", to be published, Scieuce, 1972.
[ 111 B . D . Zamarayev and A. I. Kalmykov, "On the possibility of determining the spatial structure of an agitated ocean surface by means of radar", Izvestia USSR Academy of Sciences-Atmospheric and Oceanic Physics, V O ~ . 5, pp. 64-66, 1969.
[12] G. T . Ruck, D. E. Barrick, and T. Kaliszewski, "Bistatic radar sea-state monitoring", Battelle, Columbus Labora- tories, Columbus, Ohio, Res. Rep., Contract NAS6-2006, December, 1971.
[13] L. S. Miller and G. S. Hayne, "System study of the geodetic altimeter concept", Research Triangle Institute, Research Triangle Park, North Carolina, Final Rep., Contract NAS6-1829, March, 1971.
192-OCEAN '72